First proposed human test of CRISPR passes initial safety review

The genome-editing method CRISPR may soon be tested in a clinical trial for the first time.

iStockphoto

First proposed human test of CRISPR passes initial safety review

A cancer study that would represent the first use of the red-hot gene-editing tool CRISPR in people passed a key safety review today. The proposed clinical trial, in which researchers would use CRISPR to engineer immune cells to fight cancer, won approval from the Recombinant DNA Advisory Committee (RAC) at the U.S. National Institutes of Health, a panel that has traditionally vetted the safety and ethics of gene therapy trials funded by the U.S. government and others.

Although other forms of gene editing have already been used to treat disease in people, the CRISPR trial would break new ground by modifying three different sites in the genome at once, which has not been easy to do until now. The study has also grabbed attention because—as first reported by the MIT Technology Review—tech entrepreneur Sean Parker’s new $250 million Parker Institute for Cancer Immunotherapy will fund the trial.

“It’s an important new approach. We’re going to learn a lot from this. And hopefully it will form the basis of new types of therapy,” says clinical oncologist Michael Atkins of Georgetown University in Washington, D.C., one of three RAC members who reviewed the protocol.

The proposed CRISPR trial builds off the pioneering efforts of Carl June and others at University of Pennsylvania (UPenn) to genetically modify a cancer patient’s own immune cells, specifically a class known as T cells, to treat leukemia and other cancers. For the CRISPR trial, a UPenn-led team wants to remove T cells from patients and use a harmless virus to give the cells a receptor for NY-ESO-1, a protein that is often present on certain tumors but not on most healthy cells. The modified T cells are then reinfused back into a patient and, if all goes well, attack the person’s NY-ESO-1–displaying tumors. The UPenn team has already tested this strategy in a small clinical trial for multiple myeloma. But although most patients’ tumors initially shrank, the reintroduced T cells eventually became less effective and stopped proliferating.

To boost the staying power of the engineered T cells, the UPenn group wants to use CRISPR to disrupt the gene for a protein called PD-1. The protein sits on the surface of T cells and helps dampen the activity of the cells after an immune response, but tumors have found ways to hide from T cell attack by flipping on the PD-1 switch themselves. (Drugs that block PD-1 eliminate this immune suppression and have proven to be a promising immunotherapy cancer treatment.)

June’s team also wants to knock out two gene segments that encode different portions of the protein that makes up a T cell’s primary receptor so that the engineered NS-ESO-1 receptor will be more effective. To do this, they will introduce into the T cells so-called guide RNAs, which tell CRISPR’s DNA-snipping enzyme, Cas9, where to cut the genome.

The 2-year trial will treat 18 people with myeloma, sarcoma, or melanoma who have stopped responding to existing treatments at three sites that are members of the Parker Institute—UPenn; the University of California, San Francisco; and the University of Texas MD Anderson Cancer Center in Houston. June pointed out to RAC that his team already has experience with gene editing. They have used a different technique, called zinc finger nucleases, to disrupt a gene on T cells that HIV uses to enter the cells. In a small trial, this strategy appeared to be safe and has shown promise for helping HIV patients. Those data suggest that CRISPR gene editing should be safe in humans, June said.

To confirm that, researchers conducting the CRISPR trial will look for signs of an immune reaction to the Cas9 enzyme, which comes from a bacterium. They will also look for evidence that it has made cuts in wrong places, potentially creating or triggering a cancer gene. When the UPenn team recently used CRISPR to edit T cells from healthy donors as a test run, they checked the 148 genes they most feared Cas9 would mistakenly slice and only found one cut in a harmless location. For the CRISPR trial, the team will do various tests to watch for uncontrolled growth of the modified T cells. Because they are cutting the genome in three places, one RAC member also noted, the team should watch for large swapped chunks of chromosomes.

Another concern raised by several RAC members is that June, who would not treat the cancer patients but would serve as the trial’s scientific adviser, and UPenn have a financial interest in the trial. (June has patents on using engineered T cells to treat cancer and has advised companies developing these treatments.) Some on the panel suggested they were particularly sensitive about such concerns given that it was at UPenn in 1999 that a young man, Jessie Gelsinger, died in a gene therapy trial, setting the field back for years. “Penn does have an infamous history in this regard,” says biomedical ethicist and RAC member Lainie Ross of the University of Chicago in Illinois.

However, others on the panel noted that the university could take various steps to mitigate the conflict of interest, for example by recusing June from specific tasks. UPenn itself should decide whether it can directly treat patients or merely supply the modified T cells to other sites for the trial, RAC concluded. Ultimately, RAC members voted unanimously (with one abstention) to approve the trial.

Although RAC endorsement is a big step, the researchers must now seek approval from their own institutions’ ethics boards and the U.S. Food and Drug Administration. Others are likely nipping at their heels. Many thought the Cambridge, Massachusetts–based biotech company Editas Medicine would conduct the first CRISPR clinical trial—it has announced plans to use CRISPR to treat an inherited eye disease in 2017—but RAC has not yet reviewed a proposal from the company.

*Clarification, 24 June, 10:12 a.m.: A previous version of this item incorrectly stated that CRISPR would be used to edit three genes, rather than three genome sites in two genes.